Northeast Climate Change Vulnerability Assessments - Dawson Creek submitted by: SHIFT Collaborative November 2019 - Fraser ...
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Northeast Climate Change Vulnerability Assessments FINAL REPORT Dawson Creek submitted by: SHIFT Collaborative November 2019 1
Table of Contents 1. Introduction ............................................................................................................................ 4 2. Methodology ........................................................................................................................... 8 3. How Will Climate Conditions Change in Northeastern BC?.................................................. 13 4. What Impacts Can We Expect From These Changing Conditions? ....................................... 17 5. What Impacts Are We Most Vulnerable To? ........................................................................ 18 6. Which of These Impacts Pose the Highest Risk? .................................................................. 21 7. Initial Action Ideas................................................................................................................. 26 8. Conclusion ............................................................................................................................. 30 Appendix A. Risk-based Adaptation Planning Tools ..................................................................... 31 Appendix B. Regional Results of Pre-Workshop Survey ............................................................... 35 Appendix C. Qualitative Vulnerability and Risk Assessment Process ........................................... 45 Appendix D: Impact Statements with Additional Detail ............................................................... 49 Appendix E. Heat Map .................................................................................................................. 58 Appendix F. Climate Adaptation Teams, Terms of Reference ...................................................... 61 Appendix G. Detailed Action Plan ................................................................................................. 63 Appendix H. Climate Adaptation Maturity Scale .......................................................................... 78 List of Figures Figure 1. Northeast Climate Vulnerability Assessment Project Overview ...................................................... 6 Figure 2. Assessing risk and vulnerability to changing climate conditions .................................................... 7 Figure 3. Municipal service areas with potential impacts due to changing climate conditions ................... 9 Figure 4. Three step vulnerability and consequence assessment method used in workshop 1 .................11 Figure 5. Action planning process ................................................................................................................12 Figure 6. Current levels of concern about climate-related hazards for Northeast communities ...............17 Figure 7. Survey Respondents from Across Communities...........................................................................35 Figure 8. Survey respondents understanding of climate change ................................................................37 Figure 9. Survey respondents’ agreement that their community is doing enough to adapt to climate change. ................................................................................................................................................38 Figure 10. Concerns about effects of climate change .................................................................................39 Figure 11. Frequency for considering climate change in one’s work. .........................................................41 Figure 12. Perception of need for climate change adaptation supports ....................................................42 Figure 13. Barriers to climate change adaptation .......................................................................................43 Figure 14. Perception of key stakeholders that should be involved in developing climate change adaptation measures ..........................................................................................................................44 Figure 15. Sensitivity and adaptive capacity rating scales Used in workshop 1 assessment process.........47 Figure 16. Vulnerability assessment matrix, based on BARC-ICLEI .............................................................48 Figure 17. Heat Map characterization of risk ratings for highest vulnerability impact statements. ..........58 2
List of Tables Table 1.Temperature 15 Table 2. Heating and Cooling 15 Table 3. Growing Season 16 Table 4: Precipitation 16 Table 5. Climate impact statements with highest vulnerability in Dawson Creek 18 Table 6. Summary of Climate Hazards and Risk Ratings for Dawson Creek 22 Table 7. Potential opportunities for regional collaboration across the Network 27 Table 8. Steps in Climate Change Adaptation Methodology 33 Table 9. Respondents by Sector 36 Table 10. Highest risks identified for Dawson Creek 59 3
1. Introduction The Northeastern region of BC represents 21.8% of the land area of the province (20,494,470 ha), but is the least populated region in BC, with just 1.6% of the population (about 69,068 people). The Northeastern region is expected to undergo significant population and economic growth as a result of rapid expansion in natural gas and oil development. (The population is expected to increase by 30% over the next 25 years, BC Stats, 2014). There are two broad types of actions relating to climate change: reducing greenhouse gas emissions (climate change mitigation), and preparing for and adjusting to changes in climate that impact human and non-human systems (climate change adaptation). The focus of this effort is climate change adaptation. Recent Federal research1 indicates that Canada is warming at twice the global average rate, with severe impacts already being felt and anticipated to increase in severity. Compounding this, northeastern BC has experienced a significantly more rapid rate of warming over the last 100 years (2.2 °C) than has BC overall (1.2°C). Significant changes in climate are projected for the future. The region is expected to see increases in average temperature and precipitation, the frequency and severity of extreme precipitation, drought and flood events, and other associated impacts. Changes to the climate and extreme weather is challenging Recent Climate Events in the industry and communities across the Northeast region of the Northeast Region province. Wildfires, hail storms and floods have already • Annual near flood events impacted local infrastructure and posed health risks to across region communities. Ensuring the region is as prepared as possible • 2017 Pouce Coupe Hail Storm for future climate events is critical to maintaining a thriving • 2016 Dawson Creek flood community, robust natural environment, and vibrant • 2016 Chetwynd Flood economy. “As prepared as possible” means the region • 2014 Regional Drought understands how the climate is changing and is working together to increase resiliency and improve natural and physical infrastructure. Early efforts will reduce the reliance on emergency management and support the ability to change and thrive over time. Local governments in the region are taking a proactive approach to understand how climate change will pose risks to Northeast communities and are planning together to build resiliency across the region. Designing to current and future climate parameters is anticipated to be more cost-effective than reacting to climate shocks and stresses over time. 1.1 The Northeast Climate Risk Project This project is being conducted for the Northeast Climate Risk Network (NECRN). The NECRN includes partners from: the City of Fort St. John, City of Dawson Creek, District of Tumbler Ridge, 1 Government of Canada, Natural Resources Canada https://www.nrcan.gc.ca/environment/impacts-adaptation/21177 4
District of Chetwynd, Village of Pouce Coupe and Northern Rockies Regional Municipality who have come together to better understand and build capacity to address the impacts of climate change. NECRN is a peer-mentoring network on adaptation to climate change, which also serves as an advisory body for the overall Northeast Climate Risk Project. This project was initiated in 2018 by the Fraser Basin Council, in partnership with the six local municipalities. The Project has three goals: 1. Support the Northeast BC local government partners in preparing for a changing climate and understanding the associated risks and vulnerabilities; 2. Collaboratively address climate risks at a regional and community scale through a peer network; and, 3. Increase staff and stakeholder awareness of climate change impacts through the planning process as a first step to building community, public and private sector awareness of climate change impacts. The NECRN has been working collaboratively to produce a number of products that will support communities in the region to better prepare for and adapt to a changing climate, namely: Northeast BC Regional Climate Projections Report The regional climate projections report2 provides Northeast BC with regional information on climate projections, precipitation indicators, summer and winter temperature indicators and hydrology for selected river catchments, for the 2050’s and 2080’s. Community Scoping Reports An internal gap analysis was conducted for each of the six municipalities, which identified priority climate change impacts, outlined related work done to date and identified priority gaps and opportunities to address changing climate conditions. Community Vulnerability Assessments (this document) This report is the final comprehensive report on the results of the year-long process outlined in Figure 1. Each community prioritized impacts for initial action planning employing a vulnerability and risk assessment. 2 https://www.fraserbasin.bc.ca/Northeast_BC_Climate_Risk.html 5
Northeast Climate Risk Network (NECRN) Climate Vulnerability Assessment Project 2018 - 2019 Draft Outline & Vulnerability Workplan Assessment Reports Develop a Develop draft detailed vulnerability workplan for assessment Vulnerability reports for each Assessment community, for Phase review Gap Analysis Assessing Generating and Plan & Build Final Reports Impacts, Evaluating Capacity for Vulnerability Adaptation Implementation and Risk Options Document NECRN workshop NECRN individual Community- review & (with key staff / calls and joint Develop final specific interviews à stakeholders) to videoconference vulnerability workshops to community generate initial to identify near assessment assess impacts, scoping reports ideas of term priority report for each vulnerability, risk identifying gaps adaptation actions and community, of community and options and explore with near term assets & values opportunities introduce opportunities for action plan to climate for adaptation methods for building collective change impacts prioritization capacity July Jan Mar Sept Dec 2018 2019 2019 2019 2019 Figure 1. Northeast Climate Vulnerability Assessment Project Overview 6
1.2 What determines risk and vulnerability to changes in climate? A key first step on the way to adapting to climate change is to consider what the relative risks and vulnerabilities are likely to be for the various components of the community. To do this, we have taken a commonly used approach to understanding risk and vulnerability, as represented in Figure 2: Adaptive Capacity (to what extent can the system prepare and respond?) Exposure & Sensitivity (what is Consequence (degree exposed and how sensitive is it to of impact) impact?) Risk = Likelihood (how likely Hazard (changes to climate) is this to happen?) Figure 2. Assessing risk and vulnerability to changing climate conditions The risks posed by climate change are a combination of the characteristics of the climate- induced hazard itself (e.g. magnitude of the hazard (such as flooding) and frequency) and how this interacts with human, built and ecological systems. The degree to which these systems are vulnerable, depends in part on whether they are exposed to the hazard, and how sensitive they are to its effects. For example, a housing development in a floodplain is exposed to flood risk while a subdivision at higher elevation is not. Similarly, a newer development built with floodproofing features is less sensitive than an older home without any such features. But vulnerability also depends on the ways in which the affected systems can anticipate, prepare and adapt to better manage the challenge (adaptive capacity). For example, communities with a well-developed emergency management system or who have used tools and resources to divert development to areas outside of the floodplain, are more adaptive to an increasing flood hazard. The degree of impact that occurs as a result (consequence), combined with how likely the event or impact is to occur, provides an indication of the degree of risk due to these changing conditions. 7
2. Methodology Risk management across domains follows the basic steps listed and includes an iterative process of adaptive management and continuous improvement: identification of risk, analyzing risk to understand whether it requires a response, and choosing the best risk management options3. Several adaptation or risk-based planning tools are variations on these common steps (see Appendix A). All of these approaches use relatively qualitative assessment techniques for vulnerability and risk, which is the appropriate methodology for higher-level assessments such as this one. This Vulnerability Assessment has a specific purpose of better understanding the range of impacts the community may face and prioritizing across and between them as a basis for deciding on where to focus their limited resources and time. The need for more detailed assessments may be identified as a near-term action step resulting from this prioritization process4. In order to consider the range of ways that a changing climate will impact the community and the work of the local government specifically, impacts were organized by municipal service areas, as shown in Error! Reference source not found.. These service areas were considered in the gap analysis, pre-workshop survey and throughout the vulnerability and risk assessment. This Vulnerability Assessment Report is the culmination of three key components: 1. A community-level gap analysis; 2. A qualitative vulnerability and risk assessment process with stakeholders; and 3. An action planning process led by each community’s Climate Adaptation Team. Each of these components is outlined below (and in the associated appendices), followed by a discussion of the limitations of the methodology. 3 See ISO 31000 Risk Management Framework https://www.iso.org/iso-31000-risk-management.html 4For example, an assessment of a sewerage area or a flood risk assessment can rely on damage-curves and specific calculations of consequences to provide quantitative detail where it is needed. Engineers Canada’s PIEVC protocol has a specific option to undertake a more detailed engineering study, and is a common approach to climate risk assessment for engineered infrastructure. 8
Municipal Service Areas Figure 3. Municipal service areas were used to illustrate the range of areas impacted by climate change. 2.1 Gap Analysis The gap analysis was conducted to produce the “Community Scoping Report” for Dawson Creek, a stand-alone internal document for use by the local government. The gap analysis consisted of a combination of document review and interviews. Documents were reviewed to identify: • Key background and contextual information about the community and systems that may be vulnerable to climate change impacts; • Any policies or actions that directly or indirectly address potential vulnerabilities to climate change; • Any immediate or planned windows of opportunity for incorporating a climate lens; and • Any existing information, studies, and reports that contribute to an understanding of climate change projections or vulnerabilities in the community or the northeast region. Following the initial review of documents, a number of interviews were conducted with local government staff and other agencies, to fill in gaps in information. Information from these two key sources was compiled in the resulting report. Information gaps and opportunities for 9
applying a climate lens were identified by service area (e.g. sewage and drainage, community development). 2.2 Vulnerability & Risk Assessment Process This step in the methodology aimed to outline which impacts from a changing climate pose the greatest risk to each community, and thus form the basis for prioritizing near-term actions. To better understand the baseline knowledge and perspectives of participants about the impacts and potential for adaptation to changing climate conditions, a survey was conducted of confirmed participants in advance of the workshop. The compiled results of this survey across all participants in the region, are included in Appendix B. On February 28th, 2019 a workshop was held in Dawson Creek with City staff and other community and regional stakeholders, to discuss the possible climate impacts for Dawson Creek and begin to prioritize which impacts to focus on in the near term. Eighteen people participated in this workshop, representing the following organizations: • City of Dawson Creek (12) • Peace River Regional District • Peace River Forage Association • Northern Health • Northern Environmental Action Team • BC Ministry of Forests, Lands, Natural Resource Operations and Rural Development • BC Oil and Gas Commission The workshop began with a summary of the regional climate projections outlined in the report “Climate Projections for the BC Northeast Region5.” Impact statements describing the types of impacts expected for the community (based on information from the Gap Analysis), were developed as the basis for discussion and input during the workshop. These impact statements were taken through a three-step process6, outlined in Figure 4 below: 5Produced by Pacific Climate Impacts Consortium and Pinna Sustainability, completed in 2019. 6The process used borrows from the ICLEI BARC process and employed the BARC rating scale for vulnerability. The focus for impact statement generation was the climate projections to 2050. When assessing vulnerability and risk, current community characteristics were used to simplify the process. 10
Gap Analysis + Climate ~50 Im pact ~25 Im pact Projections Statem ents Statem ents Step 1: Generate Step 2: Assess Step 3: Assess Impact Statements Vulnerability Consequence (Risk) Add, change or Provide a rating of Provide a rating of drop Impact “Sensitivity” and “consequence” Statements (initial “Adaptive across five list generated from Capacity” for each categories, for Gap Analysis). impact statement each of the higher vulnerability Provide additional Prioritize higher impacts detail to better vulnerability understand each impacts impact in this community. Figure 4. Three step vulnerability and consequence assessment method used in workshop 1 See Appendix C for further detail on the assessment process used to qualitatively assess vulnerability (sensitivity x adaptive capacity) and consequence during Workshop 1. Three University-based experts with background in the Northeast region and climate change provided likelihood scores for the impact statements. The consequence ratings for the higher vulnerability impacts generated during the workshop were combined with these likelihood ratings to provide an overall risk score. Those impact statements that fall into the medium to high risk category were then cross-referenced with the gap analysis results and reviewed by the community’s Climate Adaptation Team in Workshop 2 to arrive at the final list of priority impacts. 2.3 Initial Steps Toward Action Planning In order to operationalize the findings of this assessment process each community was invited to form a Climate Adaptation Team to attend the second workshop and lead the initial action planning. A description of the purpose, potential membership and functions of these teams is outlined in Appendix F. The final list of higher risk impacts was used as the basis for action planning. Action planning focuses on exploring options and next steps to reduce these key risks and enhance resilience to these climate impacts. Lower risk impacts were provided to the teams as well, to scan for those that could easily be addressed by low-effort/no-regret actions. The action planning process was introduced in Workshop 2, and then completed by each Climate Adaptation Team working with the support of the project team. Each community’s action plan was shared with NECRN 11
members, who together identified opportunities for regional collaboration and learning. This process is outlined in Figure 5. Ground-truth Review and provide amendments to the Assessment list & heat-map of top risks for each Results community. Generate & Practice working with evaluation Practice Evaluating criteria and designing for co-benefits. Adaptation Generate possible actions for top risks. Options Working in Climate Adaptation Teams Complete an Initial and with other staff & stakeholders as Action Plan needed, complete the action plan template. Identify Opportunities for As a network, review community action plans and identify opportunities Regional for joint action and learning. Collaboration Figure 5. Action planning process 2.4 Limitations of the Methodology Efforts have been made at each step of the process to “ground-truth” results with those who have local knowledge and/or topic-specific expertise. However, there will be inevitable gaps and omissions due to the breadth of what is being considered. This report is to be considered a working document – as communities take the next steps to address priority impact areas, more information will emerge and be developed that will inform further steps. Adaptation assessment and planning is an inherently iterative process as understanding and conditions (i.e. the climate, political will, priorities, land use patterns, infrastructure needs, innovations, etc.) develop over time. Due to time and availability, the number of interviews for each community during the Gap Analysis was constrained and this limits the range or depth of perspectives that we were able to gather on specific aspects of system vulnerability. While additional information and perspectives were incorporated through stakeholder feedback on the impact statements in Workshop 1, this was still limited by the knowledge of participants who were able to attend. Numerous adaptation methodologies (i.e. see Appendix A for examples) rely on stakeholder opinions as a basis for vulnerability and risk assessment. That said, the qualitative approach to 12
risk and vulnerability assessment has a number of limitations. Ratings for sensitivity, adaptive capacity and consequence were all based on informed judgment of stakeholders who were present for Workshop 1. As such, there are gaps in the expertise and understanding of the relationships between future climate conditions and how these would affect various parts of the system. This was mitigated by having multiple people rate each impact statement and deliberate on the final rating assigned to each, along with two additional steps to double-check the results with participants and a further step of cross-referencing with information compiled during the Gap Analysis. Still, these are subjective ratings and may contain biases, for example: 1) Recently experienced impacts may be rated higher in vulnerability and/or consequence 2) More familiar impacts may be rated differently than less familiar impacts 3) Impacts of general importance to participants may be rated more highly than others (e.g. someone who works on water treatment may rate associated risks more highly than those to emergency volunteer capacity) It is important to remember that the purpose of this assessment is to provide a basis for prioritizing more important impacts for the purposes of allocating time and resources to addressing these. This is a first step, providing a high-level assessment that supports the community to make more informed choices about how to prepare and adapt to the projected changes in climate conditions. Likelihood was rated by four academics with firsthand research experience in the Northeast region, in related fields. Their ratings were generally quite similar but not identical for all impacts, and were averaged where they did not converge. 3. How Will Climate Conditions Change in Northeastern BC? Dawson Creek is considered a lowland community based on elevation. Dawson Creek and the other network communities can expect significant changes to its climate in the coming decades, including: • Increased precipitation across all seasons. The largest increases in precipitation will take place during the spring & autumn months during which, on average, the region can expect 30% more precipitation by the 2050s. This can lead to more frequent flooding and stress to ecosystems. • Summers will be considerably warmer. In the past, the region experienced an average of 12 days over 25°C annually. We can expect 32 days by the 2050s (49 days by the 2080s). Dawson Creek is expected to experience 40 days above 25°C by the 2050s and 60 days by the 2080s, with 10 days above 30°C by the 2050s and 21 days over 30°C by late century. By the 2080s, summer temperatures are projected to be about as warm as Kelowna’s past summers (1980s). 13
• Summer is expected to remain the wettest season, though by a smaller margin. Even though precipitation is projected to increase slightly over summer months, on average, normal seasonal variability in precipitation plus hotter temperatures (and thus increased evaporation) could lead to drier, hotter summers, posing increased risk of wildfire and associated physical and mental health impacts. • Temperatures projected will trigger significant heat stress across the region. The ability to provide clean drinking water as a shared resource to communities, industry and agriculture may be strained. Higher demand for water during longer, hotter summers as well as during dry spells, could create challenges for water supply, water quality, livestock and crop yields. • Winter temperatures are also projected to warm. By the 2080s, January temperatures are projected to feel like March temperatures of the past, with warmer nights, 28% less frost days and 37% longer growing seasons than the past. More winter precipitation and later onset of freezing temperatures could potentially lead to additional annual freeze thaw cycles and more frequent rain-on-snow events. Shifting seasonal temperatures could also result in premature pollination of crops and increased invasive species and pests, impacting agriculture and forestry. • More extreme storm events in the future. As the climate warms, more moisture is held in the atmosphere, resulting in more intense precipitation during extreme events. Future storms may also bring stronger winds and hail events. These events will challenge regional infrastructure, and may overwhelm sewerage and drainage systems. Other extreme weather events of concern include river floods, high winds, drought, and wildfire. • Summer streamflow will decrease in all basins. Warmer temperatures mean that relatively more precipitation will fall as rain rather than snow, which in general means an increase in winter runoff, reduced snowpack and an earlier freshet for the Kiskatinaw River. Less water stored over winter and melted earlier in the year also means reduced summer and early fall streamflow for the Kiskatinaw and other area rivers. These changes will be exacerbated by increased evaporation, further contributing to increased stress to water resources. 14
The projected changes to climate conditions for Dawson Creek and Northeastern BC7, are summarized in the tables below. Table 1.Temperature Expect summers and winters to warm with summer days getting hotter Description of Metric Anticipated change for NEBC Lowlands Annual average temperature increase Average Annual temperature will increase +3.7°C by 2050 and +6°C by 2080s Overall Change: Summers get warmer with hotter “hot days” Number of days where maximum temperature is Number of summer days increases by 50% above 25°C (from 16 up to 24) Maximum daytime temp of the warmest summer Warmest summer day is 3.7°C warmer day Maximum night time temperature in the summer Hottest summer night is 3.3°C warmer Hot summer days = days above 30°C in any one year The number of summer days above 30°C is projected to occur 9 times/year by the 2050s and 20 times/year by 2080s. (previously occurred ~1 time/year) Maximum daily temp expected to occur on average Very hot days increase in intensity from 1:20 years (5% chance in any year that the temp 31°C to 38°C by the 2080s could reach this magnitude). Overall Change: Winters get warmer, especially nights Min temp of the coldest night time in winter Coldest winter nights: 6.5°C warmer Max temp of the warmest night time in winter Warmest winter night: 2.4°C warmer Minimum daytime temperature of the coldest day in Coldest winter day: 6.2°C warmer winter Max daytime temperature of the warmest day in Warmest winter day: 1.5°C warmer winter Minimum night time temp expected to occur on Very cold nights are projected to warm average 1:20 years (5% chance in any year that the from -44°C to -38°C temp could reach this magnitude). Ice Days is an annual count of days when daytime 17% fewer ice days high temp is less than 0°C Table 2. Heating and Cooling Greater need for summer cooling while winter heating demand decreases slightly Description of Metric Anticipated change for NEBC Lowlands Indicator for heating demand. 18% fewer Heating Degree Days (Total of the number of degrees below 18°C that occur daily, summed over each 7 Data as generated by Pacific Climate Impacts Consortium (www.pacificclimate.org) in the Regional Climate Projections Report. All projections are to the year 2050 unless noted. Change is relative to the baseline period of 1971 – 2000. All projections are averages from Global Climate Models. 15
day of the year) Indicator for cooling demand. 600% more Cooling Degree Days (18 – 94 days (Total of the number of degrees above lowlands) 18°C that occur daily, summed over each day of the year) Table 3. Growing Season Longer growing season with a doubling of Growing Degree Days (GDD) Description of Metric Anticipated change for NEBC Lowlands Growing season length is the length Growing season length increases from 160 to 189 between the first span of six days above days in 2050, and 205 days in 2080. 5°C in spring, and the first span of six days below 5°C in the fall. Annual count of days when daily minimum 17% fewer frost days temp is below 0°C GDD is an indicator for plant growth: Lower elevations can expect 2000 GDD by 2080s Total of the number of degrees above 5°C (81% increase from the past) that occur daily, summed over each day of the year. Table 4: Precipitation Expect more precipitation annually with the largest increases in the spring and autumn. Summer remains the wettest season to the 2050s. Expect an increase in frequency and intensity of rain events. Description of Metric Anticipated change for NEBC Lowlands Average seasonal changes in precipitation 19% increase in spring precipitation, 15% increase in autumn, 7% in summer and 11% in winter Overall Change: Intensity (volume/ size of event) of significant rain events will increase Wettest day is the largest amount of rain 17% increase (from 28mm) in amount of rain on that falls on any single day in the year on the wettest day of the year average Largest amount of precipitation that falls 19% increase in rain amount over 5 consecutive over a period of 5 consecutive days in the days to 68mm by the 2080s year. Overall Change: More frequent intense precipitation events and more rain to fall during these events. Annual total precipitation that falls on the The amount of precipitation falling on very wet wettest days of the year (days where days of the year, increases by 35% by 2050 and 16
precipitation exceeds the 95th percentile 51% by 2080. amount recorded during the baseline period). As above but 99th percentile The amount of precipitation falling on the wettest days of the year, increases by 59% by 2050 and 78% by 2080 (from 35mm). Max daily precipitation expected to occur A 1:20 year return precipitation event will on average once in 20 years (5% chance in increase in intensity by 25% (to 61mm of any given year) precipitation) in a day 4. What Impacts Can We Expect From These Changing Conditions? Workshop 1 participants from each community were invited to complete a survey (see Appendix B) that asked about climate-related hazards of concern for their community (Figure 6). The top concern across all communities was wildfire (78% of respondents ranked this as something they are most concerned about), followed by flooding (56%), other extreme events such as storms, hail, freezing rain (53%) and extreme rainfall events (51%). The climate-related hazard communities were least concerned about was extreme heat (20%). 90% 78% 80% Level of Concern Across All 70% 60% 56% Communities 53% 51% 50% 40% 30% 20% 20% 10% 0% Wildfire Flooding Other extreme Extreme rainfall Extreme heat events (storms, hail, freezing rain) Figure 6. Current levels of concern about climate-related hazards for Northeast communities In terms of the impact that changing climate conditions could have on municipal service areas, respondents from Dawson Creek were equally concerned about a number of municipal services—67% of respondents were concerned about negative impacts to each of: water supply, storm drainage, transportation and hazard and emergency management. 17
An important step in this assessment was to translate knowledge about the projected future changes in climate-related hazards into the impacts of concern to municipal service areas. Through the gap analysis and Workshop 1, information was gathered to create impact statements that clearly define the impact a changing climate would have on important aspects of community life. Impact statements combine specific climate hazard features (e.g. increasing potential for rain-on-snow events) with specific impacts to community functioning, assets and/or values (e.g. flooding, street closures and mobility issues). An example of an impact statement is: “Increasing potential for rain-on-snow events (and freezing rain) increases the risk of street flooding, street closures, mobility issues and potential damage to nearby buildings.” There is a wide range of potential impacts to municipal service areas of the changing climate conditions in the Northeast region – a detailed list is included in Appendix D, which provides additional details relating to each impact statement. The full list of potential impacts to municipal service areas was prioritized through the vulnerability and risk assessment process, to identify those impacts of greatest concern and where attention should be prioritized in the near term. Those impacts are outlined in Sections 5 and 6 below. 5. What Impacts Are We Most Vulnerable To? As described above (see Figure 2), vulnerability is a combination of exposure (whether an element is exposed to the climate hazard), sensitivity (how sensitive the element is to its effects) and adaptive capacity (the ways in which the affected systems can anticipate, prepare and adapt to better manage the challenge). Based on the assessment completed in workshop 1, the following impacts (listed in Table 5 below) were identified as those the community is most vulnerable to, given current adaptive capacity. Table 5. Climate impact statements with highest vulnerability in Dawson Creek Climate Hazard Impact Statement • Increased river flow and flood events damage buildings, crops, parkland, infrastructure such as bridges, etc. • Increased river flow and flood events cause increased avulsions (changes to the river’s path) and bank erosion, impacting aquatic habitat. • More frequent flood events increase costs and require more capacity for flood clean-up and restoration • More frequent flood events contribute to cumulative anxiety, stress and mental health challenges among affected residents and service providers 18
• Increased flood events cause strain and stress on municipal wastewater systems, and private layouts and dugouts (also private septic and lagoons) • Warmer winter temperatures may enable spruce beetle infestations which decreases local forestry incomes • More rapid snow melt leads to increasing localized flooding from backed up culverts • Increasing freeze/thaw cycles decrease the durability of infrastructure (such as roads and trails) and increase maintenance needs • Increasing potential for rain-on-snow events (and ice storms) increases the risk of street flooding, street closures, mobility issues and potential damage to nearby buildings • Increased presence of ticks • Rain or snow events increase run-off in non-typical times of the year (January). Catalyst for localized flooding • Situations of increased turbidity due to larger or more frequent precipitation events decrease the number of days that pumping can occur from the river to water supply reservoirs • Heavy rainfall events cause more sewer back-up, which damage buildings and can cause health impacts • Increasing surface run-off during the rainy season and decreasing recharge during the longer dry spells will impact groundwater and base flows • Increasing intensity of rainfall causes more frequent localized flooding and sewer surcharge in parts of the community • Increased surface runoff can lead to increased nutrient concentrations in the water and increased sediment load thereby impacting operation of water supply infrastructure • Increase in extreme events cause more frequent power outages, in turn affecting household power and power to critical infrastructure • Increasing frequency and intensity of extreme events puts pressure on the physical and emotional capacity of community volunteers including ESS and volunteer fire fighters • Increasing extreme events decreases the durability of infrastructure generally, leading to earlier replacement and increased maintenance needs • More frequent / larger scale emergency and disaster events place (i.e. storms, hail, increasing demands on municipal resources and capacity for extreme heat, mitigation, planning, preparedness, response and recovery wildfires, landslides) • Increasing number or magnitude of wildfires, impacts the health of the community due to poor air quality days and mental health 19
impacts • Major events (landslide, flood, wildfire) lead to impacted transportation within the community or between communities. (i.e. isolation from critical services, jobs, evacuation routes) • The combination of hotter and drier conditions increases potential for drought and decreased water supply in late summer – industry is first to be impacted when conservation measures are introduced In Summary: Flooding was identified as a major vulnerability for Dawson Creek including overland river flooding and urban flooding. River and creek flooding may cause damage to infrastructure, crops, public and private waste water systems as well as the environment. These impacts all require capacity and resources to clean up and recover and impacted residents and responders are vulnerable to anxiety and other mental health effects. As the climate changes, increasing rain on snow events and freeze/thaw cycles are anticipated and the municipalities more vulnerable to urban flooding from blocked sewers and culverts etc. Transportation, mobility and exposed infrastructure are all vulnerable to urban flooding. With warmer winters, an increase in pests is anticipated. Economic vulnerability to agriculture pests and the spruce beetle were identified as well as health vulnerability to ticks. Water supply vulnerabilities were identified from a range of climate changes. The infrastructure is vulnerable to increasing turbidity and sediment load resulting from heavier rainfall. Drier summers and heavier rains may impact groundwater and flow volume making water management planning more challenging. Industry may be vulnerable to water supply pressures felt late in the summer. Vulnerabilities to extreme events range from health impacts, to the capacity to respond and recover to lifeline impacts such as power outages and transportation disruptions. 20
6. Which of These Impacts Pose the Highest Risk? As Figure 2 depicts, risk is a product of consequence and likelihood. The degree of impact that occurs as a result of the climate hazard (consequence), combined with how likely this is to happen, provides an indication of the degree of risk due to these changing conditions. Risk results for Dawson Creek fall into the high to medium-low range. None of the impacts were assessed as low risk. To be rated high or medium-high risk, the impact has a combination of high consequence rating and medium or high likelihood rating, high ratings for both, or a medium consequence rating and high likelihood rating. Table 6, below summarizes the risk assessment results, also indicating which service areas are most affected by each impact statement. 21
Table 6. Summary of Climate Hazards and Risk Ratings for Dawson Creek 22
23
24
25
For Dawson Creek, flood impacts related to the Creek flooding and intense precipitation events featured prominently among the 18 highest risks identified (Table 6) with the impacts of extreme events and drier periods also having important effects. A range of types of flood impacts made up the majority of the top risks for Dawson Creek, with concern about impacts due to both localized surface flooding (rainfall/rain on snow related) and riverine (rivers overcoming capacity) flood events. Flood impacts will be an important focus for the next steps of this work. Aside from the impacts of floods, there were a handful of additional key risks. The effects of both flood and other extreme events, and changes to freeze-thaw cycles, were of concern with respect to the durability of infrastructure. Impacts to water supply and quality due to changes in precipitation patterns and dry / drought conditions were important. The impact of extreme events was another key issue, with concern for issues such as disrupted transportation and mobility within and between communities (and isolation from critical services); the stress and capacity drain for staff, volunteers and first responders; and effects of power outages. The impact of spruce beetle infestations on forestry incomes rounded out the top risks. 7. Initial Action Ideas Community-specific In September 2019, the project team from Dawson Creek put together initial action ideas for addressing the highest risk impacts. For each higher risk impact the team began developing: • A list of actions, • A description of the action, • The status of the action (underway, act now / by X year, monitor or investigate further), • The range of funding and HR to carry out the action (low, medium or high), • Next steps, and • Notes on how well the action addresses evaluation criteria To view the initial action ideas for Dawson Creek, please visit Appendix G. Detailed Action Plan. Region-wide Adapting effectively and building resilience to a changing climate will require action at both a local and regional level. Actions that lend themselves to regional collaboration were discussed with the Northeast Climate Risk Network throughout the course of this project. Broadly across hazards, the Network provides opportunities to: • share purchasing power for hard goods or common consultancy needs; • advocate as a united front with stakeholders and other levels of government; 26
• avoid costly projects if recommendations or best practices from one municipality can be transferred to another; • share progress and discuss barriers and solutions and; • partner as a group on common projects and work with the regional governments and other stakeholders. Hazard-specific opportunities for regional collaboration are summarized in Table 7 below. Table 7. Potential opportunities for regional collaboration across the Network Climate Hazard Impact Statement • Chetwynd and Dawson Creek have parallel processes currently including identifying and studying the hazard, mapping flood extents, coming up with mitigation options and including findings in the OCP. While at different steps, this provides an opportunity to learn from each process and pass lessons on across BC. • The region has to work together and work with the Province on understanding forest management for pest (and wildfire) resilience • Snow and ice clearing policies and road and bridge closure policies came up frequently as requiring updates and clear communication to the public. Sharing policies and best practices makes sense. • Steaming vs. Heat tracing culverts. A study across municipalities and conditions would provide more fruitful conclusions and recommendations. • PRRD Everbridge Emergency Notification System collaboration o Freezing rain information and health and social wellness information is shared on this platform already • Could intentionally set up to trial different types of materials to use on various roads during temperature fluctuations. • Invasive species programs provide another opportunity to learn from each other and possibly collectively hire consultants. • There are several sewage and drainage studies underway in the region. Resultant recommendations could be shared with the network via webinar with consultants on hand. • Understanding how to incorporate new rainfall projections and IDF curves into design could be done by a common consultant or process shared by one municipality to the others. • Several municipalities are further ahead on green infrastructure use (for infiltrating rainfall). Sharing actions and any pamphlets produced for the public is an opportunity for collaboration. • Dawson Creek has significant expertise in watershed and riparian health and may focus more on this area. Best practices could be 27
shared through the Network meetings. • Set up a more formal contact person and liaison between the Network and the Ministry of Transportation and Infrastructure. They have done significant work on climate change guidelines and the Network municipalities identified flooding, landslides and wildfires to have the potential to limit access to their communities. • Several communities will refresh or initiate landslide and steep slope reviews. This provides an opportunity to share consultants or at least process and best practices. Translating findings into land use controls may also be an area to share ideas. (i.e. storms, hail, • Asset management becomes an important element of ensuring extreme heat, infrastructure is resilient to climate change. Opportunity: wildfires, landslides) collaborating on a project to apply a climate lens to asset management (in one area or across the board). • Air quality monitoring is already completed by Oil and Gas industry. Collaboration with industry could add air quality data for rural and urban areas. • Municipalities have Community wildfire plans at different stages of development, update and implementation. Opportunity to learn from each other and also possibly share a consultant from out of town for economies of scale. • Fuel removal communication / education campaign could be shared • Network could partner with the regional health authority to have consistent messaging on wildfire smoke, heat and health • Municipalities all work on water conservation techniques for drought. Many deal with agriculture and/or industry water needs as flow volumes change, and as a result, there are lessons to be shared. Key Competencies for Advancing Adaptation As part of the Gap Analysis at the beginning of this project, the consultant team provided an initial assessment of Dawson Creek’s stage of development around climate change adaptation planning, using the Climate Adaptation Maturity Scale developed by the Federation of Canadian Municipalities. The checklist version used to guide the assessment, is included in Appendix H. The scale has three competency areas: policy, human resources and governance, and technical and risk management capacity. Each competency area has 5 levels through which a municipality can progress. The competencies are: • Policy: Putting in place policies and objectives related to the development of an environment and vision that supports local climate adaptation. 28
• Human resources and governance: Ensuring staff and council are equipped with the mandate, understanding, skills and knowledge needed to support local climate adaptation. • Technical and risk management capacity: Preparing the tools needed to deliver adaptation initiatives and manage operations in a way that minimizes climate risk (e.g. software, hardware, maps, models, etc.). At the beginning of this project, Dawson Creek was considered to be at Level 1 for two out of three competencies (Human Resources and Governance, and Technical and Risk Management Capacity), and working towards Level 2 for Policy. At the end of the action planning process, the assessment using the Maturity Scale was completed again. Based on this assessment, Dawson Creek is working on level 2 for Human Resources and Governance as they build awareness of the work completed during this project with decision-makers and leaders across the organization. Level 2 for Policy will be completed at the end of this project. On Technical and Risk Management Capacity, completing this project has moved the dial from level 1 to level 3 and for some impacts to level 4 for Dawson Creek. Staff have been committed to assessing risks, forming a team and brainstorming adaptation responses. The organization has already put significant resources into planning for climate- related hazards such as flooding and this project has complemented those efforts. Next Steps The Climate Adaptation Team is well-positioned – with an overall understanding of the climate projections, potential impacts and priority risks – to continue working across disciplines in the organization and in collaboration with community stakeholders to mitigate risks and build resilience to changes in climatic conditions moving forward. The initial action ideas included above and in Appendix G provide direction and focus in these next phases of the work. An important aspect of implementation is continual monitoring and review. With climate change, more so than some other things that we plan for, conditions are highly complex and will continue to change and evolve over time. As a result, we need to continue to monitor changes to: 1. Climate projections – particularly any regional and local data; 2. Actual and potential impacts – for example, air quality due to wildfire was not a common concern just a few years ago. New impacts will emerge and existing ones will evolve; 3. Opportunities to mitigate risks and build resilience – windows of opportunity should be constantly in mind to take advantage of sources of funding, partnerships, awareness, and political will, etc. 29
It is recommended that the Climate Adaptation Team develop a monitoring and review strategy to ensure that this function is continually addressed. This may include: • A review of the climate science every 3-5 years • A scan of existing and emerging impacts every 1-2 years • Re-evaluating the top priority impacts and action ideas every 2-3 years As well, the NECRN is a valuable resource for all communities in the region that will require ongoing participation and investment of time and energy by its members to cultivate that value. The Fraser Basin Council will be convening the network in January of 2020, creating an opportunity to identify some initial areas of collaboration and joint projects. Generating some early wins and getting some experience in working together across communities would be a helpful foundation for the work moving forward. 8. Conclusion The six communities in the Northeast Climate Risk Network have all experienced climate- related shocks (such as a flood or wildfire) or a stress (such as drier conditions, or more rain-on- snow events in the winter) in recent years. These events have impacted city services, community assets and residents. The climate vulnerability assessments have built on the good work started in each community to prepare and respond to a changing climate. Based on information provided in an initial community-specific Gap Analysis and the climate projections for the region for the time periods of the 2050s and 2080s, staff and stakeholders from the communities and the region identified impacts to a range of service areas. The vulnerability of certain assets or areas was assessed and risk ratings supported prioritization of the impacts to ensure that resources can be directed to where they are most needed. Throughout the process there were opportunities to learn from and share with others across the network, and to develop capacity within communities to continue with this work. Within each community, cross-disciplinary staff teams considered the top risks for their community as they identified and developed action ideas to mitigate risks, and to prepare and respond to the climate change-related impacts. These teams are well-placed to continue to guide and lead this work forward into implementation and ongoing monitoring and iterations. A range of possible ways to work together, capitalize on resources across communities, and maintain and build capacity as a region were identified. Together, these assessment results and the initial action ideas developed by the Climate Adaptation Teams reflect the growing capacity of each community and the Northeast Climate Risk network as a whole to form a strong foundation for building resilience across the Northeast Region. 30
Appendix A. Risk-based Adaptation Planning Tools Infrastructure or risk-specific assessments, such as an assessment of a sewerage area or a coastal flood risk assessment can rely on damage-curves and specific calculations of consequences to provide quantitative detail where it is needed. Engineers Canada’s PIEVC protocol has a specific option to undertake a more detailed engineering study—the need for a more detailed assessment like this is something that may be identified through the more qualitative risk assessment. This higher level risk assessment has a specific purpose of better understanding the range of impacts the community may face, and prioritizing across and between them as a basis for deciding on where to focus their limited resources and time. The risk-based land use guide for Canada outlines steps for planning including: • establish the community needs, • define hazard potential, • identify what could be harmed, • calculate potential losses (consequences), • recommend safe land-use, • and monitor decision success. The UKCIP tool (Adaptation Wizard) for planning looks at current climate vulnerability and future climate vulnerability (risk included) before brainstorming adaptation options and monitoring and reviewing iteratively. The ICLEI BARC tool follows 5 milestones including: • initiating (taking a first look at impacts and adaptation actions underway; • researching climate science and completing vulnerability and risk assessments; • planning (identifying and evaluating actions); • implementation and; • monitoring and review. The ICLEI BARC tool relies on clear impact statements throughout the vulnerability and risk assessments. The International Standards Association (ISO) published a standard for climate change adaptation (ISO 14090) in June, 2019. The standard provides a framework to help organizations consider climate change adaptation in policies, strategies and plans. A companion standard (ISO 14091) is under development currently and will specify risk and vulnerability assessment steps. In July, 2019 the provincial government published the Preliminary Strategic Climate Risk Assessment for British Columbia. The publication includes a strategic climate risk assessment framework for B.C that was developed to be scalable to analyze risks at multiple levels and a variety of climate-related risks. The framework has 4 steps: • Understand the Context (including scope and objectives and audience); • Identify risk events (they use a scenario approach to risk events); 31
You can also read